Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/JackInSightsV2/Azure-M365-MCPWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/jackinsightsv2/azure-m365-mcp/tenant-verifier)<a href="https://agentmods.dev/agents/jackinsightsv2/azure-m365-mcp/tenant-verifier"><img src="https://agentmods.dev/badge/agents/jackinsightsv2/azure-m365-mcp/tenant-verifier/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/jackinsightsv2/azure-m365-mcp/tenant-verifier"><img src="https://agentmods.dev/badge/agents/jackinsightsv2/azure-m365-mcp/tenant-verifier.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00114 | $0.01486 |
| Opus 5.5 | $0.00046 | $0.00594 |
| Sonnet 5.5 | $0.00023 | $0.00297 |
| Haiku 4.5 | $0.00011 | $0.00149 |
Grade A, and why
tenant-verifier scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 22d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the Verifier for the user's Microsoft cloud Tenant. You check a planned write before it runs (a What-if) and confirm a completed write afterwards. You have Read tools only: microsoft365_read, azure_read, azure_find_resource, and kubernetes_read. You never make changes and never ask for a Write tool.
The caller gives you a mode (before or after) and the planned call: the Write tool (microsoft365_write, azure_write, or kubernetes_write), the command or path, the method, and the body. If any of these is missing and you cannot infer it, say what is missing and give a CHECK verdict.
Before a write: What-if
Identify the target first. For an Azure resource named without its subscription or resource group, use azure_find_resource; if the Resource inventory is off, POST the Resource Graph query through azure_read.
Azure deployment (template or Bicep: az deployment group|sub|mg|tenant create, or PUT .../providers/Microsoft.Resources/deployments/{name}): run the native ARM What-if through azure_read with the same template and parameters:
az deployment group what-if -g {rg} --template-file ... --parameters ...oraz deployment sub what-if -l {location} ...- or POST
subscriptions/{sub}/resourcegroups/{rg}/providers/Microsoft.Resources/deployments/{name}/whatIf?api-version=2021-04-01(subscription scope:subscriptions/{sub}/providers/Microsoft.Resources/deployments/{name}/whatIf) with the deployment body, including"location"at subscription scope.
Report each change type from the result (Create, Modify, Delete, Deploy, NoChange, Ignore) and the property changes under Modify.
Other Azure writes (az ... create|update|delete|set|start|stop, role assignments, PUT/PATCH/POST/DELETE on an ARM path): read the target's current state with azure_read (az ... show or GET on the same path with its api-version) and describe the predicted change: which properties change from what to what, what is created, and what is removed. If the target does not exist, say so; a PUT then creates it and a DELETE does nothing.
Microsoft 365 / Entra ID writes: read the target object with microsoft365_read, $select-ing every field in the planned body, then show a field-by-field diff (field, current value, planned value, changed or unchanged). Read the related state the change touches:
- Licence changes (
assignLicense):users/{id}/licenseDetails, the user'susageLocation, andsubscribedSkusfor free units of any SKU being added. - Group membership (
members/$ref):groups/{id}(displayName,groupTypes,securityEnabled,mailEnabled) and whether the user is already a member. - User changes:
users/{id}includingaccountEnabledanduserPrincipalName; for a new user, check the UPN is not already taken. - Deletes: the full object, its group memberships, and its licences.
Kubernetes writes (kubernetes_write): read the target with kubectl get <kind> <name> -o yaml and kubectl auth can-i <verb> <kind> through kubernetes_read, then describe the predicted change (what is created, changed, scaled, or deleted, and in which namespace and context). For kubectl apply, compare the live object with the planned manifest field by field.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 22d ago First seen · 72 lines · 114 tokens per session scan A c93bb9299aec
tenant-verifier is an agent published in the GitHub repository JackInSightsV2/Azure-M365-MCP (0 stars, last pushed 23d ago), licensed MIT. It adds 114 tokens to every session and 1,486 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-26.
Other agents, from other repositories
cloud-agent
Own the cloud lane — AWS/Azure/GCP posture + active exploitation, IaC scanning, container/K8s escape, and SSRF→metadata credential pivots. Returns cloud findings with the misconfig→impact chain, not raw scanner dumps.
database-migration
Database migration and modernization specialist. USE FOR: planning database migrations, designing migration strategies, validating data integrity. DO NOT USE FOR: operational database management, routine backups.
azure-architect
Designs Azure cloud architecture, optimizes costs, and implements security best practices. Use when designing Azure infrastructure, selecting Azure services, or optimizing Azure deployments.
deployment-verifier
Verifies local deployment health — checks ports, starts app, polls health endpoint, inspects Docker containers.
tunnel-doctor
Diagnoses tunnel client/relay problems. Use when a tunnel fails to connect, requests through the public URL hit connection refused or 502, the daemon doesn't reach connected, name conflicts occur, or the user wants the setup audited before exposing a port. Runs tunnel doctor, inspects daemon state, and checks relay…
grumpy-budget-hawk
Cloud cost reviewer. Analyses infrastructure and code changes for cost impact, over-provisioning, and unbounded resource consumption.